Scrippsiella acuminata versus Scrippsiella ramonii: A Physiological Comparison.

Scrippsiella is a cosmopolitan dinoflagellate genus that is able to form Harmful Algal Blooms in coastal waters. The large physiological, morphological, and genetic variability that characterizes this genus suggest the existence of cryptic species. In this study, flow cytometric analyses were carried out to compare the cell cycle and life cycle of two Scrippsiella strains from two different species: Scrippsiella ramonii (VGO1053) and Scrippsiella acuminata (S3V). Both species were also investigated by internally transcribed spacer rDNA sequencing and high-performance liquid chromatography-based pigment analyses. The reddish-brown color of S. acuminata and yellowish-green hue of S. ramonii were consistent with the quantitative differences determined in their pigment profiles. Our results indicate that the cell cycle is light-controlled and that it differs in the two species. S-phase was detected during the light period in both, whereas the G2/M phase occurred during the light period in S. ramonii but under dark conditions in S. acuminata. The detection of 4C stages, mobile zygotes (planozygotes), and resting cysts in S. ramonii (nonclonal) provided convincing evidence of sexuality in this species. Sexual related processes were not found in the clonal S. acuminata strain, suggesting its heterothallic behavior (i.e., the need for outcrossing). The differences in the genome size of these species were examined as well.

Dinoflagellates are unicellular microalgae that belong to the larger group of eukaryotic, typically photoautotrophic microorganisms. While the morphology, physiology, biochemistry, and ecology of dinoflagellates are extremely diverse, genomic analyses have proven to be particularly challenging 1, such that only one genome has been sequenced so far, and only partially 2. The dinoflagellate nuclear genome is typically large, often >250 Gb, which is 70 times the size of the human genome 3. Other complexities and peculiarities of the dinoflagellate genome are its permanently condensed liquid‐crystalline chromosomes, the lack of histones, frequent base‐pair substitutions and a large number of repetitive sequences.

Dinoflagellate species use multiple adaptive strategies to survive and grow under a wide variety of environmental conditions 4, 5. They notoriously form harmful algal blooms (HABs) in coastal areas around the world. HABs reflect the rapid proliferation and/or the high biomass accumulation of toxic or otherwise noxious microalgae at the sea surface or in the water column.

The high‐density HABs formed by Scrippsiella, a genus of nontoxic, cosmopolitan marine dinoflagellates, lead to oxygen depletion and therefore fish kills 6. Scrippsiella blooms have been reported worldwide, including in China 7, 8, 9, 10, Ukraine 11, Spain 12, the southern Red Sea 13, Mediterranean Sea 14, 15, the Atlantic and Pacific coasts of Mexico 16, 17, 18, 19, Australia 6, the Atlantic coast of the USA 20, and the southeast coast of Iran 21. Although little is known about the physiology of this genus, its large physiological, morphological, and genetic variability suggest the existence of cryptic species. In fact, Scrippsiella trochoidea, the most common species of the genus, is a species complex that consists of a cryptic diversity and multiple species 22. Moreover, true S. trochoidea must be considered a heterotypic synonym of S. acuminata 23, this last name having priority over S. trochoidea (Stein 1883) as it comes from Peridinium acuminatum, described by Ehrenberg (1836). However, a species redefinition based on molecular phylogeny requires specific phenotypic analyses in order to correctly classify and characterize S. acuminata. The physiological comparison conducted in the present study revealed relevant, previously unexplored differences between two Scrippsiella species, S. acuminata and S. ramonii, with respect to their genome size, photosynthetic pigments, and life cycle.

As in all eukaryotic cells, dinoflagellate growth (mitotic cycle) involves DNA replication and yields two genetically identical daughter cells. Replication in eukaryotes can be divided into five stages: (1) interphase or G1, during which the vegetative cells (with a given “C” DNA content) grow and accumulate nutrients; (2) S‐phase, when DNA is replicated; (3) G2, from the completion of DNA replication (2C) until the beginning of cell division; (4) mitosis, or karyokinesis, during which the nucleus divides; and (5) cytokinesis, or cytoplasmic division. However, to study dinoflagellate growth, its life cycle transitions must also be taken into account, as sexuality involves a change in the cellular DNA content. With the exception of Noctiluca, dinoflagellates have a haplontic life cycle, in which mitotic division occurs only in the haploid stage and the zygote is the only diploid (2 N) stage 24. Under certain environmental conditions, many species initiate a sexual cycle, during which vegetative cells differentiate into gametes that fuse to form diploid zygotes, which can either divide or form resting cysts that are deposited in the benthos (see the review by Figueroa et al. 25). Consequently, a 2C DNA content represents either a vegetative G2 cell or a diploid (2 N, zygote), arising from the fusion of two gametes (C, N). Given that morphological differences between vegetative and sexual stages are very small or undetectable (reviewed by Bravo and Figueroa 26), studies of the cell ploidy state are essential to identify sexual events in dinoflagellates. When, to what extent and by which route sexuality is conducted are still poorly understood. Moreover, in addition to the complex patterns of the dinoflagellate life cycle, there are numerous species‐specific peculiarities. It is therefore clear that bloom formation by dinoflagellates and the ability of these organisms to adapt to a wide variety of environmental conditions cannot be fully understood without investigations of their life cycle 4, 5. For example, for some species a sexual cycle has never been described, whereas for others a resting stage has never been found or it lacks a sexual origin 27. Therefore, dinoflagellate life‐cycle must be studied individually in order to fully understand each species environmental response.

Among the approaches used to study the life cycle transitions of dinoflagellates are the isolation and monitoring of individual cells and determinations of the changes in the DNA content of cell populations using flow cytometry. HPLC‐based pigment analyses provide chemotaxonomical fingerprint that can be associated with other phenotypic traits (e.g., changes in color and the corresponding physiological changes). Flow cytometry detects and quantifies multiple optical properties; its applications have included lipid quantification 28 and the estimation of genome size 29, 30. Image flow cytometry (IFC) is a recently introduced technique that combines the fluorescence sensitivity of standard flow cytometry with the spatial resolution and quantitative morphology of digital microscopy 31. It has been used in analyses of the dinoflagellate cell cycle and in the identification of dinoflagellate life cycle stages (e.g., Ref. 32). In the present study, we used flow cytometry, cell sorting, and IFC to examine the mitotic cycle and to determine whether a sexual cycle is a feature of S. acuminata and S. ramonii.

Methods

Experimental Organisms and Culture Conditions

The dinoflagellate species used in this study were Scrippsiella ramonii (strain VGO1053) and Scrippsiella acuminata (strain S3V). Both are regularly maintained at the culture collection of the Centro Oceanográfico de Vigo (CCVIEO; Culture Collection of Harmful Microalgae of the Spanish Institute of Oceanography). The strains were cultured at 25 ± 1°C with an irradiance of approximately 90 μmol photons m−2 s−1 and a photoperiod of 12:12 h light:dark (L:D), with the light period starting at 8:00 and ending at 20:00. The cells were grown and maintained in L1 medium 33 without added silica. The medium was prepared with NE Atlantic seawater and adjusted to a salinity of 32 psu by the addition of sterile double‐distilled water.

DNA Analyses

Molecular analyses

DNA extraction, PCR amplification, and sequencing

Samples for the molecular analyses of both Scrippsiella strains were processed as follows. Exponentially growing cultures (1.5 ml) were harvested by centrifugation (15,871g, 2 min) using an Eppendorf 5424R centrifuge (Eppendorf AG, NY). The cell pellets were rinsed in 1 ml of distilled water, centrifuged again and the supernatants discarded. DNA was extracted from samples frozen overnight at −20°C according to either (1) a modified Chelex procedure, following the method of Fraga and Rodríguez 34 or (2) a CTAB protocol adapted from Doyle 35.

DNA extracts from Scrippsiella were quantified and checked for purity using a Nanodrop Lite spectrophotometer (ThermoScientific, Waltham, MA). The D1–D2 regions of the LSU rRNA gene were amplified using the primer pairs D1R/D2C (5′‐ACCCGCTGAATTTAAGCATA‐3′/5′‐ACGAACGATTTGCACGTCAG‐3′) 36. Internally transcribed spacer (ITS)‐1/5.8S/ITS‐2 rDNA regions were amplified using the primer pair ITSF01/PERK‐ITS‐AS (5′‐GAGGAAGGAGAAGTCGTAACAAGG‐3′/5′‐GCTTACTTATATGCTTAAATTCAG‐3′) 37. The amplification reaction mixtures (20 μl) contained 10 μl of Taq DNA polymerase master mix (2×) (Canvax Biotech, Córdoba, Spain), 0.5 pmol of each primer, and 1–2 μl of DNA extract.

The regions of interest were amplified in an Eppendorf Mastercycler EP5345 (Eppendorf AG) as follows: 5 min of denaturation at 94°C, followed by 35 cycles of 35 s of denaturation at 94°C, a 35‐s annealing step at 50°C and 1 min of elongation at 72°C, and a final elongation step of 7 min at 72°C.

The integrity of the DNA was confirmed by checking a 9‐μl aliquot of each PCR sample by agarose gel electrophoresis (1.5% TAE, 80 V) and SYBR Safe DNA gel staining (Invitrogen, CA). The PCR products were purified with ExoSAP‐IT (USB Corp., OH), sequenced using the Big Dye Terminator v3.1 reaction cycle sequencing kit (Applied Biosystems, Foster City, CA) and migrated in an AB 3130 sequencer (Applied Biosystems) at the CACTI sequencing facilities (Universidade de Vigo, Spain). The amplified ITS and LSU rRNA gene sequences obtained in this study were deposited in GenBank.

Phylogenetic analyses

The obtained ITS rDNA sequences were used in the phylogenetic determinations, in which sequences of Scrippsiella ramonii (VGO1053, 599 nt) and S. acuminata (S3V, 616 nt) were compared with the sequences of 30 other Scrippsiella spp. strains obtained from GenBank. A sequence from Pentapharsodinium tyrrhenicum was used as the outgroup to root the tree. ITS (ITS‐1/5.8S/ITS‐2 regions) sequences were aligned using BioEdit v.7.2.5 (final alignment: 567 nt). Phylogenetic model selections were performed on MEGA 7, with the TN93 + G substitution model selected for both Scrippsiella phylogenies (gamma shape parameter, γ = 0.05). The phylogenetic relationships were also determined using Bayesian phylogenetic inference, using the substitution models obtained by sampling across the entire general time reversible model space as described in the Mr. Bayes v3.2 manual. The Bayesian tree was produced using Mr. Bayes v3.2 38 and the program parameters statefreqpr = dirichlet (1,1,1,1), nst = mixed, rates = gamma. One million generations were used in the analyses. Posterior probabilities were calculated from every 1,000th tree, sampled after a log‐likelihood stabilization (“burn‐in” phase). Maximum likelihood (ML) phylogenetic analyses were conducted in MEGA 7. Bootstrap values were estimated from 1,000 replicates. The overall topologies obtained using the ML and Bayesian inference methods were very similar. The phylogenetic tree was represented using the ML method, with bootstrap values and posterior probabilities from the Bayesian inference.

Cell cycle

Sampling

Scrippsiella ramonii: Cultures of S. ramonii that had reached 8,000–10,000 cells ml−1 (corresponding to the early exponential growth phase) were synchronized by placing them in complete darkness for 66 h, as described by Taroncher‐Oldenburg et al. 39 and Figueroa et al. 40. Light conditions were then restored and the cells were inoculated into fresh L1 medium to a concentration of approximately 3,000 cells ml−1 (1.3 l in total). Sampling started 24 h after inoculation, to allow cell recovery and the initiation of cell division. An automatic water sampler (AWS, EnviroTech Instruments, Chesapeake, VA) was used to collect 24‐ml samples every 2 h for 60 h, with 1 min of programmed gentle magnetic shaking just before sampling. For the determination of absolute cell numbers (cell concentration analyses), a 3‐ml sample was collected manually each day of sampling, fixed with Lugol, and counted in an inverted microscope at 100× magnification using a Sedgwick‐rafter chamber.

Scrippsiella acuminata: The prolonged darkness during synchronization resulted in irreversible damage to growing cells of this species. However, preliminary tests showed that this species is naturally synchronized, with few or no S‐phase cells at the beginning of the light period. Therefore, S. acuminata cultures were not synchronized. When the cultures reached an approximate concentration of 10,000 cells ml−1 during a light:dark incubation (12:12), 3,000 cells were inoculated into fresh L1 medium, for a total culture volume of 1.6 l. Sampling started 24 h later, during which 25‐ml samples were collected manually every 2 h for 60 h, with shaking of the cultures just before sampling. For the determination of absolute cell numbers, two 3‐ml samples, the first at 11:00 and the second at 15:00, were collected manually each day, fixed in Lugol and counted as described above for S. ramonii.

Flow cytometry analyses and cell sorting

Samples for cell cycle and genome size studies were DNA stained and analyzed by flow cytometry. Because preliminary tests showed that Scrippsiella cells were highly resistant to DNA staining, the samples were first permeabilized by fixation for at least 10 min with 1% formaldehyde in a mixture with 1% “L permeabilization solution” 41. The samples were centrifuged at 9,000g for 10 min for the genome size analysis, and filtered through a 5.0‐μm‐pore‐size membrane filter (Millipore, Ireland) for the cell cycle analyses. Following Kremp and Parrow 27, the retained cells were resuspended in 1 ml of a fixation solution (3:1 methanol:glacial acetic acid + 10% DMSO + 0.1% Triton‐X 100) and stored for at least 12 h at 4°C to facilitate chlorophyll extraction. The cells were centrifuged at 9,000g for 10 min and then washed in phosphate‐buffered saline (PBS) (pH 7, Sigma‐Aldrich, St. Louis, MO). The pellet was resuspended in 0.33 ml of staining solution [60 μg propidium iodide ml−1 (Sigma‐Aldrich) in PBS and 100 mg RNaseA ml−1 (Sigma‐Aldrich) in distilled water] and maintained in darkness for at least 3 h before the analysis.

Statistical analyses, including those yielding the graphs displayed in Figure 4, were performed using the statistical and programming software R 2.1.12 (R Development Core Team, 2012) with the packages “ggplot2” and “scales,” available through the CRAN repository (www.r-project.org/).

Figure 4

Comparisons between Scrippsiella ramonii (upper panel, A) and Scrippsiella acuminata (lower panel, B) cell cycles. Percentage of cells in each cell cycle stage as shown by a trend line and the standard error (gray‐shaded area). The dark bars indicate the dark period.

Cell sorting by flow cytometry

A SH800Z cell sorter (Sony Biotechnology Inc., Europe, United Kingdom) equipped with a 488‐nm diode laser was used to analyze and sort the samples. The samples were run at medium‐high pressure and data were acquired until 10,000–30,000 events in the gated population had been recorded. Each cell cycle phase was delimited by means of a histogram of propidium iodide fluorescence. A617/30 emission filter was used for propidium iodide detection [FL3 (PI)]. According to their DNA content and nuclear fluorescence, the cells were assigned to the following peaks: 1C, S (C ➔ 2C), 2C and 4C, in which 1C is the amount of DNA in the unreplicated gametic nucleus of an organism. ModFit LT 4.0 (Verity Software House, Topsham, ME) was used in the cell cycle analyses to calculate both the percentages of cells in the DNA fluorescence peaks and their ratios. The software FlowJo 7.6 (Tree Star, Inc.) was used in the genome size analysis to compute peak numbers, coefficients of variation (CVs), and peak ratios for the DNA fluorescence distributions in a population. Runs with CVs > 13.4 were discarded from the analyses.

Cell sorting was conducted at medium speed and in purity mode. Sorted cells with peak DNA contents of 1C, 2C, and 4C were stained with calcofluor (0.5 mg ml−1) and imaged at 1,000× magnification (Leica DMR; Germany) using a microscope camera (Axiocam HRC, Zeiss Germany).

Imaging flow cytometry

A Flow Sight image flow cytometer (Amnis, Seattle, WA) was used to study the morphology of the cells in different phases of the cell cycle. The software Ideas 6.0 (Amnis) was used to process the images. The image‐processing methodology followed that described in Dapena et al. 42, such that only images in focus were selected while those depicting artifacts and aggregates were discarded.

Growth parameters

The specific growth rate 43 was estimated as K′ = , where N1 and N2 are the cells counts at times t1 and t2 of the exponential growth phase.

Divisions per day were calculated as Div day−1 = .

Genome size

Duplicate exponentially growing cultures from each species were used to determine and compare the genome sizes of S. acuminata and S. ramonii. The manually collected 25‐ml samples were gently mixed just before sampling at the beginning of the light period (9:00). This time point was chosen to obtain a maximum number of cells in the1C peak, used for genome size estimations, before its widening due to the onset of S‐phase. The samples were stained and analyzed as described for the cell cycle analyses. Genome size was estimated by comparison with the genome sizes of two Alexandrium minutum strains, AMP4 and VGO577, which according to Stüken et al. 29 have an average genome size of 26.2 ± 3 and 25.7 ± 3.2 pg, respectively.

Pigment Analyses

High‐performance liquid chromatography (HPLC) was used to determine pigment composition in exponentially growing cultures. The ∼40‐ml samples were filtered through glass microfiber filters (0.7‐μm pore‐size, Filter‐Lab, Spain) and stored at −20°C until use.

The pigments were extracted by the addition of 3 ml of 90% acetone to the frozen filters. After a 15‐min incubation in the dark at 4°C, the filters were ground, sonicated for 5 min in an ultrasonic bath filled with a mixture of water and ice and then centrifuged (3,500g, 5 min). The supernatant was filtered through hydrophilic PTFE membrane filters (0.22‐μm pore size) to separate the extract from the filter remnants and cell debris. The HPLC samples were prepared by mixing 130 μl of extract with 75 μl of Milli‐Q water in the autosampler loop. Pigments were separated following Zapata et al. 44. The Waters Alliance HPLC system (Waters Corp.) consisted of a 2,695 separations module and a Waters 996 diode‐array detector (1.2‐nm optical resolution). Pigments were identified by co‐chromatography with authentic standards obtained from SCORreference cultures and by diode‐array spectroscopy (see Zapata et al. 44).

Results

Phylogenetic Analysis

The phylogenetic tree obtained for S. ramonii strain VGO1053 and S. acuminata strain S3V based on their respective ITS‐1/5.8S/ITS‐2 rDNA regions is shown in Figure 1. It confirmed that strain VGO1053 (acc. no. MH359388) belonged to S. ramonii, as the ITS‐1/5.8S/ITS‐2 rDNA sequence was identical to that of S. ramonii from the Mediterranean Sea (acc. no. HQ729497). Similarly, strain S3V was confirmed as S. acuminata, based on the identity of its ITS‐1/5.8S/ITS‐2 rDNA sequence (acc. no. MH359389) to that of another S. acuminata sequence (acc. no. KJ189493) positioned within a subclade of S. acuminata (STR2, as described in Kretschmann et al. 45).

Figure 1

Phylogenetic trees inferred from ITS sequences based on maximum likelihood (ML). Pentapharsodinium tyrrhenicum was used as outgroup. Numbers on branches are statistical support values (Bayesian posterior probability/ML bootstrap support). Only bootstrap values >60% and posterior probabilities of 0.6 or above are shown. Hyphens indicate bootstrap/posterior probability values <60/<6. New sequences obtained in this study were highlighted in gray (Scrippsiella acuminata MH359389, S. ramonii MH359388). PRE: Clade of Scrippsiella precaria and its relatives; STR: Clade of Scrippsiella acuminata and its relatives; subclades SPI, STR1, STR2, and STR3. PRE and STR were labeled following Luo et al. (2016). Scale bar indicates number of nucleotide substitutions per site.

Cell Cycle

Scrippsiella ramonii (strain VGO1053)

Growth

The growth rates of the cultures were calculated during every light:dark period. From Day 1 to Day 2 of the cell cycle, the growth rate was 0.78 ± 0.05 divisions day−1 whereas from Day 2 to Day 3 it was higher, 0.89 ± 0.04 divisions day−1.

DNA content

The nuclear fluorescence histogram (Fig. 2) revealed two main peaks in the DNA content of S. ramonii cells, at 1C and 2C (2C/1C = 1.9–2).

Figure 2

Cell cycle of S. ramonii (left) and S. acuminata (right). Flow histograms showing DNA intensity obtained every 2 h for the studied light:dark periods. Blue peak showed G1 (1C cells), red peak showed G2 (2C cells), and green color is the S‐phase (1C → 2C) (Modfit LT 4.0 analysis). Yellow squares indicate light period and gray squares indicate dark period. Events number: 5,000–30,000.

The main peak was at 1C, where the percentage of cells was between 29% and 42%. However, at the end of the light period, in the 17:00 and 19:00 samples, the 1C peak decreased and the 2C peak increased to 56%–65% of the population. The percentage of the population in the 2C peak never declined below 20%.

S‐phase was detected only during the light period, with a maximum (19%) reached during the middle of this period and a minimum (5%) at the beginning of the dark period.

During the first L:D period, a 4C peak was also observed that reached a maximum of 4% at 19:00. The detection of a 4C stage indicated the existence of a sexual process in the culture, which was also evident by the detection of resting cyst production. In addition, the IFC images (Fig. 3, image 9,867) demonstrated the existence of planozygotes, identified by their double longitudinal flagella (arrows), whereas flow cytometric cell sorting showed that the 4C cells were neither artifacts nor aggregates (Fig. 6F). The IFC images also revealed differences between the nuclear morphologies of cell populations with different DNA content peaks. Thus, 1C cells were characterized by a small, round nucleus (Fig. 3, images 79 and 87), whereas 2C cells were seen as single cells with either elongated (Fig. 3, images 1,110 and 1,309) or round (Fig. 3, images 1,317 and 9,867) nuclei or as cells in a two‐celled chain (Fig. 3, images 949 and 8,633). The round nuclei of 2C cells were larger than the single nuclei of 1C cells. Resting cysts (2C content) were also identified based on their round shape and double external wall (Fig. 3, images 2,440 and 2,557).

Figure 3

IFC pictures of Scrippsiella ramonii. IFC images of cell and nuclear morphologies using bright field (BF) microscopy and after blue (488 nm) and violet (405 nm) laser excitation. Picture 9,867 shows a planozygote with the characteristic longitudinal biflagellation.

Figure 6

High power magnification (1,000×) images of Scrippsiella acuminata sorted cells at different cell cycle stages. A–C showed “1C” cells and D–F showed “2C” cells. DNA IP stains (red) and calcofluor (blue). Scale bars = 10 μm.

Scrippsiella acuminata (strain S3V)

The growth rate of S. acuminata was similar to that of S. ramonii albeit slower during the second L:D period. Between Days 1 and 2 of the cell cycle, the growth rate was 0.91 ± 0.06 divisions day−1 and between Days 2 and 3 0.71 ± 0.2 divisions day−1.

As in strain VGO1053, the prevailing peak in all strain S3V samples (except number 5, corresponding to 21:00) was at 1C, in which the percentage of C cells was between 37% and 98% (Fig. 2). The histograms from cells at the beginning of the light period (9:00, 11:00, and 13:00 samples) showed one main peak, corresponding to a 1C DNA content. However, as the cells progressed through the light period, an additional peak appeared, corresponding to a 2C DNA content (2C/1C ≥ 1.9). The maximum percentage of cells at the 2C stage was coincident with the detection of the lowest number of cells in 1C. S‐phase cells were detected mostly during light conditions, from 13:00 onward.

The 2C and S‐phase peaks were remarkably smaller on Day 3, which indicated that the percentage of the population remaining at the 1C stage was much higher (>68%).

A comparison of the cell cycles of S. ramonii and S. acuminata based on the percentages of cells per peak during the two light–dark periods is provided in Figure 4. In both species, the main peak during the three L:D periods was at 1C, with the percentages always >29%, while in S‐phase the highest abundances occurred during the light period. By contrast, the two species differed in the timing of their 1C and 2C DNA content stages, In the case of S. ramonii the maximum of the 1C peak was during the dark period and for the 2C peak during light period. By contrast, for S. acuminata, the maximum 1C value was during the light period and the 2C maximum in the dark period. In addition, a 4C stage was not observed in S. acuminata.

Genome Size Comparison

Haploid DNA content was calculated by comparing the DNA content of the two strains with that of the dinoflagellate Alexandrium minutum, which was previously calculated by Stüken et al. 29 and thus served as a control in this study. Those authors reported that the genome of A. minutum strain AMP4 is slightly larger than that of A. minutum strain VGO577 (Fig. S1).

The haploid DNA content of S. ramonii and A. minutum was similar, 31.88 ± 0.38 pg cell−1 (Fig. S1) whereas S. acuminata had a much larger genome, with a DNA content almost double (average of 59.64 ± 2.06 pg cell−1) that of the S. ramonii and A. minutum genomes.

Nuclear Morphologies

Images of the sorted cells, captured using epifluorescence microscopy at 1,000× magnification, are shown in Figures 5 and 6. Similar to the results obtained at lower magnification with IFC, round nuclei were typical of many 1C (Figs. 5A,B and 6A–C), 2C (Fig. 6E,F), and 4C DNA (Fig. 5F) cells. However, epifluorescence allowed more detailed observations of the nuclear morphologies of S. ramonii 2C cells at different stages of karyokinesis (Fig. 5C,D). The main difference between the nuclei of the S. ramonii and S. acuminata was the larger nuclei of the latter at all stages of the cell cycle.

Figure 5

High power magnification images (1,000×) of Scrippsiella ramonii sorted cells at different cell cycle stages. A and B showed “1C” cells, C–E showed “2C” cells, and F showed a “4C” cell. DNA IP stains (red) and calcofluor (blue). Scale bars = 10 μm.

In addition to nuclear staining, the thecae of the sorted cells were stained with calcofluor to examine the relationship between nuclear morphology and thecal shape. In contrast to the differences in DNA content and nuclear size, the presence of a very thin cell wall without thecal tabulation, instead of the typical thecal pattern, was seen in the 2C dividing cells of both S. ramonii and S. acuminata (Figs. 5E and 6D). This finding suggested the occurrence of cell division after shedding of the theca and complete “de novo” formation of the theca by the daughter cells (eleuteroschisis).

Pigment Composition

Cultures of both species showed very different coloration, as S. acuminata (S3V) was reddish‐brown while S. ramonii (VGO1053) had a yellowish‐green hue. However, HPLC analyses showed that the same pigments occurred in the two species (Fig. S2), but that the relative proportions of the major carotenoids differed. The main accessory pigments identified in both species were peridininol, magnesium 2,4‐divinylpheoporphyrin a5 monomethyl ester (MgDVP), chlorophyll c 2, peridinin, diadinoxanthin, dinoxanthin, diatoxanthin, and β,β‐carotene (Fig. S2). Of these, peridinin was the main carotenoid and its molar ratio to Chl a was almost twice as high in S. acuminata as in S. ramonii, whereas the molar ratios of diadinoxanthin, diatoxanthin, and β,β‐carotene were higher in S. ramonii (Table S3). The total pigment per cell was also higher in S. acuminata, and the peridinin content was threefold higher than that of S. ramonii.

Discussion

Scrippsiella Balech ex Loeblich III (Calciodinellaceae, “Peridiniales”) comprises a group of thecated marine, phototrophic, dinoflagellate species producing (mostly) calcareous cysts. The taxonomy of the Scrippsiella genus is still unclear and the number of species has probably been underestimated 10. In our study of the different physiological parameters of S. acuminata and S. ramonii, novel information was obtained using tools allowing the discrimination of these two Scrippsiella species.

Phylogenetic Analyses

The results of our phylogenetic analyses of S. acuminata and S. ramonii based on their ITS and LSU rDNA sequences are consistent with those previously reported for other Scrippsiella species 22, 46. Thus, S. ramonii strain VGO1053 was shown to belong to a molecular clade (“PRE” clade 47) that includes S. precaria, Montresor & Zingone, and its relatives. Strain S3V is located in the “Scrippsiella trochoidea complex” clade, which comprises morphologically indistinguishable (i.e., cryptic) species differentiated thus far only by genetic means 22, 45, 47, 48. On the basis of its ITS rDNA sequences, this species was first divided into three genetic clades, STR1, STR2, and STR3 22. However, S. trochoidea strains from the type locality could be ascribed only to the STR2 clade, which was then considered as that of “true” S. trochoidea. Later, “true” S. trochoidea was designated as the heterotypic synonym of Scrippsiella acuminata (Ehrenb.) Kretschmann, Elbr., Zinssmeister, S. Soehner, Kirsch, Kusber & Gottschling 23. For these reasons, given that in the present study strain S3V was grouped in the STR2 clade, it was identified as Scrippsiella acuminata, comb. nov. (Thoracosphaeraceae, Peridiniales).

Genome Size

The dinoflagellate genome can be enormous, ranging from ∼1.5 pg in species of the coral symbiont Symbiodinium 49, 50 to 225 pg in Prorocentrum micans 51. Our results positioned Scrippsiella in the middle of this distribution, with approximately 32 and 60 pg per haploid cell for S. ramonii and S. acuminata, respectively. These values demonstrate the variability between dinoflagellate species of the same genus, as reported for the Alexandrium genus 30, and even between strains of the same species, as shown for Alexandrium minutum strains 29. Genome size estimates are a useful tool for comparative taxonomic studies because species identity, chromosome number, and genome size are closely associated characteristics 52. To our knowledge, this is the first report of the estimated genome sizes of S. acuminata and S. ramonii.

Cell Cycle and Life Cycle

The mitotic ratio of a given dinoflagellate population determines its division rate and, consequently, its capacity to form HABs. Another key element essential in predicting the magnitude, duration, and timing of a bloom is knowledge of the life cycle strategy of the bloom‐forming species 53. Our analysis of the cell cycle and life cycle of S. ramonii and S. acuminata contributes to determinations of the similarities and differences in the growth and life cycles patterns of these species.

The cell cycle of both Scrippsiella species showed a circadian rhythm, as reported for other phytoplankton species 54, 55, 56. Thus, the cell cycle was light‐controlled over a 24‐h period, with S‐phase usually occurring during the light period, although at different hours depending on the species, as also shown for Amphidinium operculatum 57, Karenia brevis 58, 59, Alexandrium fundyense 39, and Alexandrium minutum 32, 42. In Lingulodinium polyedra 60 and Protoceratium reticulatum 61, by contrast, S‐phase occurs during the dark phase. The timing of mitosis, the G2/M phase, is also variable and species‐specific. This phase occurred at the end of the light period in S. ramonii and during the beginning of dark conditions in S. acuminata, as in A. minutum 42 and K. brevis 58.

During the cell cycle, mitosis returns cells in the 2C state (DNA replicated state, G2) to the 1C state (G1). While this observed in S. acuminata strain (S3V), it was not entirely true for S. ramonii (VGO1053), in which a remnant 2C phase always >20% was present in all samples of the cell cycle analysis. These 2C cells may have been either G2 cells that did not finish mitosis during a given light:dark period or sexual diploid cells. The latter option was supported by the detection of mobile zygotes (planozygotes), a life stage that either divides, but with a lower division rate than that of vegetative cells, or encysts 62, 63. Alternatively, planozygotes may be able to divide mitotically, as suggested for other dinoflagellate species 32, 64.

Life cycle

According to the most widespread model of the dinoflagellate life cycle, sexuality is rare and its only purpose is to create resting cysts. However, this model is now under scrutiny, as sexuality was subsequently shown to be a common phenomenon in natural blooms 65, 66, 67. Additionally, in numerous genera, including Scrippsiella, the division of mobile zygotes without encystment was reported 32, 66, 68, 69, 70. Additionally, the resting cysts of S. hangoei resting cysts were shown to be mostly asexual 27.

Our data confirmed the existence of sexuality in a nonclonal strain (VGO1053) of S. ramonii, as previously documented for this species in its original description 71. This conclusion is based on the morphological and flow cytometric characterization of a non‐sexually induced and exponentially growing population, which showed the existence of biflagellated planozygotes. Additionally, the detection of 4C cells in the flow cytometric analyses evidenced zygote division 32. Sexuality in both homothallic and heterothallic strains of S. acuminata has been reported 48. Our results suggest that strain S3V is heterothallic, based on the apparent absence in this clonal strain of planozygotes, 4C cells, or resting cysts.

Although the S. acuminata strain used in this study was clonal, the S. ramonii strain was not, but their cell and life cycles could still be compared. Vegetative division (mitosis) is independent of the mating type and the combined use of flow cytometry and morphological analyses together with observations of resting cyst production enabled sexual studies and comparisons of the life cycles of these two strains. Moreover, sexual stages (gametes and zygotes) are also produced in some clonal‐heterothallic strains of dinoflagellate species, which are thus referred to as “homozygotic,” as heterothallism is defined solely by the production of resting cysts 25. Therefore, even in clonal‐heterothallic cultures there could be gametes in the 1C peak and zygotes in the 2C peak. Mobile sexual life‐cycle stages of dinoflagellates have hardly been studied because they cannot be identified morphologically. Therefore, we identified and estimated the presence of mobile zygotes by combining a cell cycle analysis using conventional flow cytometry with morphological parameters determined using IFC and cell sorting. Nonetheless, it was still not possible to differentiate gametes, either morphologically or based on their cell cycle attributes.

Pigments

Photosynthetic pigments can be collectively used as chemotaxonomic signatures to study dinoflagellate populations, both in the field and in culture 72. The detailed study of pigment composition explained the different coloring of both Scrippsiella species. It is well known that members of the Scrippsiella genus contain peridinin‐containing plastids, which originated from secondary endosymbiosis 73, 74. At the genus level, the pigment composition of different strains or species is often identical, as was the case in the present work in which the same pigments were detected in S. acuminata and S. ramonii. However, peridinin ratios normalized to Chl a differed between the reddish S. acuminata and the greenish S. ramonii, despite identical culture conditions. These differences are in accordance with previous studies demonstrating quantitative changes in major accessory compounds (e.g., Ref. 74). Peridinin is a deep red carotenoid and the differences in its proportions relative to Chl a would account for the differences between the two species. The larger proportion of photoprotective carotenoids (diadinoxanthin and diatoxanthin) in S. acuminata than in S. ramonii indicated that irradiance acted as a stress trigger in the former (i.e., dissipation of excess excitation energy by means of the xanthophyll cycle 75).

To conclude, in this study flow cytometric analyses were effectively applied to discriminate under laboratory conditions between cryptic dinoflagellate species belonging to the same genus. Despite the morphological similarity of S. acuminata and S. ramonii, the genome of the former was shown to be twofold larger. Additionally, the mitotic cycle was characterized by a species‐specific difference in the timing of DNA replication (S‐phase). Sexuality was reported in S. ramonii during growth, being to our knowledge the first time in which mobile zygotes of this species have been described and photographed and reported to divide. Although the S. acuminata strain employed in the present study was clonal, sexuality was not observed. However, it is not possible to infer that the species is strictly heterothallic, as many more strains need to be evaluated in order to be able to establish the mating type, in case, unlike in other species, it is not strain‐dependent 25. Lastly, our physiological study of the two species was complemented by determinations of their pigment profiles, which revealed additional differences between the two species and can be used to obtain additional insights into their behavior and ecological niche.

Figure S1. Flow histograms showing 1C DNA content peak of A. minutum VGO577 (dark blue), A. minutum AMP4 (blue), S. acuminata S3V (red) and S. ramonii (green).

Figure S2. Chromatographic profiles. Pigments from cultures of S. ramonii (VGO1053, left panel) and S. acuminata (S3V, right panel). Peak labels as follows: Peridininol, Magnesium 2,4‐divinylpheoporphyrin a5 monomethyl ester (Mg‐DVP), Chlorophyll c 2 (Chl c 2), Chlorophyll a (Chl a), Peridinin, Diadinoxanthin (Diadino), Dinoxanthin (Dino), Diatoxanthin (Diato), and β,β‐Carotene (ββCar).

S3. Pigment ratios to Chl a (mol:mol−1) and pigment per cell content (fmol cell−1) in the studied strains.

Click here for additional data file.